JPH11184671A - Method, device and system for presenting information - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an information presenting device capable of flexibly controlling the presentation style of electronic information to be presented in voice. SOLUTION: A control pattern file 18 and a voice recognition dictionary 19 are composed of a control pattern file group and a voice recognition dictionary group prepared for every information unit related to a presentation style of electronic information to be presented. When the electronic information in an information presentation processing part 13 becomes an object of voice recognition control, recognition processing of an inputted voice is performed at a voice recognizing part 15 based on the voice recognition dictionary 19. At a voice control part 16, a control pattern corresponding to a vocabulary recognized on the basis of the control pattern file 18 is detected. The information presentation processing part 13 presents the corresponding electronic information in the presentation style based on the control information of the relevant control pattern.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a technique for presenting digitized information using a speech recognition technique.

[0002]

2. Description of the Related Art Conventionally, using a computer terminal or the like,
There is known a method of performing a speech recognition process based on a preset recognition vocabulary and controlling information presented on an information browsing device based on control information corresponding to the speech recognition result.

FIG. 11 is a functional block diagram showing this type of information browsing system. This information browsing system 110
Is configured to include an information providing server 30 and a plurality of clients 40 used as information browsing devices. The information providing server 30 has a database 17 built therein. The database 17 stores documents or image data to be browsed (hereinafter referred to as browse target data). Here, the means for acquiring data corresponding to the browsing request from the client 40 in the information providing server 30 and the means for controlling the communication between the information providing server 30 and the client 40 in the information providing server 30 are well-known technologies, and are omitted here. I have.

On the other hand, the client 40 has a voice recognition unit 1
11, a speech recognition dictionary 112, a speech recognition control unit 113, and a browser 114.
The voice recognition unit 111 performs a process of recognizing voice input from a user or the like via a voice input device such as a microphone. This recognition processing is performed based on the speech recognition dictionary 112 in which a vocabulary group for speech recognition is stored. The voice recognition control unit 113 controls the corresponding information on the browser 114 based on the control information corresponding to the result of the voice recognition. This control information is set in advance. The browser 114 is an information browsing interface for data provided from the information providing server 30.
The information on the browser is to be controlled by voice.

In this information browsing system 110, a user or the like browses desired information by inputting a voice control command to the client 40 and controlling information on the browser 114.

[0006]

By the way, in the above-mentioned information browsing system, when the number of words to be recognized increases, if the current speech recognition technology is applied as it is, the recognition rate will be reduced. However, there was a problem that it was difficult to achieve.

[0007] In addition, since the vocabulary to be recognized must be registered and set in advance in the speech recognition dictionary, in the information browsing system in which the information provider and the information viewer are separated, the information provider It was not possible to change the speech recognition dictionary from the side. For this reason, for example, in the control by speech recognition of information on a browser, there is a restriction that the control must be performed based on a preset speech recognition dictionary, and free control for each specific information unit such as a presentation page is not possible. Met.

SUMMARY OF THE INVENTION It is an object of the present invention to provide an improved information presenting method for realizing a form of presenting information to be presented to a browser or the like by flexible control using voice. Another object of the present invention is to provide an information presenting apparatus and system suitable for implementing the above information presenting method.

[0009]

An information presenting method according to the present invention for solving the above-mentioned problems is an information presenting method using a computer device, which changes the presentation form of digitized information to be presented by voice, A recognition vocabulary group to be subjected to voice recognition and a presentation control information group related to the presentation of the digitized information corresponding to the recognized vocabulary group are held in advance for each presentation form of the digitized information, and the input speech is stored. A detection vocabulary to be matched with the recognized vocabulary group is detected from the group of recognized vocabularies, presentation control information to be matched with the recognized vocabulary is detected from the group of presentation control information, and an electronic corresponding to the presentation form based on the detected presentation control information. Is characterized by presenting coded information.

According to another aspect of the present invention, there is provided an information presenting apparatus for controlling the presentation form of digitized information to be presented by voice. A speech recognition dictionary registered in advance for each presentation form, a control pattern file in which presentation control information relating to presentation of the digitized information corresponding to the recognized vocabulary group is set in advance for each presentation form, and information from an operator. Means for presenting the digitized information corresponding to the presentation request and detecting whether or not the digitized information is a presentation control target by voice; and A voice recognition unit that obtains a voice from an operator, detects a recognition vocabulary to be compared with the voice from the voice recognition dictionary, and performs voice recognition, and performs matching with the detected recognition vocabulary. With a presentation control information detected from said control pattern file to, and provided with a presentation means for presenting the corresponding electronic information presentation form based on the presentation control information.

In this information presentation device, the voice recognition dictionary searches for unique data corresponding to a plurality of voice recognition data having the same data format that can be compared with voices from the operator as a recognition vocabulary. Is built on. Alternatively, it is constructed such that a unique character string corresponding to a plurality of pieces of voice recognition data having the same data format that can be collated with the voice from the operator is retrieved as a recognition vocabulary.

Further, the control pattern file is constructed so that the presentation control information corresponding to the control pattern composed of the recognized vocabulary or a combination of the recognized vocabulary can be retrieved. The voice recognition dictionary and the control pattern file may be constructed in a database storing the digitized information, for example.

The presenting means is configured to present the digitized information based on, for example, an information unit corresponding to the digitized information presentation area.

[0014] Further, the information presentation system of the present invention for solving the above-mentioned other problems is a computerized information to be presented, a speech recognition vocabulary group to be subjected to voice recognition, and the electronic corresponding to the speech recognition vocabulary group. Device that previously creates and holds a presentation control information group related to the presentation of digitized information for each presentation form of the digitized information, and a second device that presents the digitized information corresponding to an information presentation request from an operator. The device is connected so that bidirectional communication is possible.

In this information presentation system, the second
The device acquires and presents the digitized information corresponding to the information presentation request from the operator from the first device, and detects whether the digitized information is a presentation control target by voice, and When the digitized information to be subjected to the presentation control by the voice is detected, the voice from the operator is obtained, and the recognition vocabulary group corresponding to the digitized information is obtained from the first device, and the voice is obtained. Means for detecting a recognition vocabulary to be compared, and presentation control information for acquiring the presentation control information group corresponding to the digitized information to be subjected to the presentation control from the first device, and detecting the presentation control information to be matched with the detected recognition vocabulary And a presentation means for acquiring and presenting the corresponding digitized information in the presentation form based on the presentation control information from the first device, and presenting the presented form of the digitized information to a voice. And controlling Ri. The presenting unit is configured to present the digitized information based on, for example, an information unit corresponding to a presentation area of a predetermined Web browser.

[0016]

Embodiments of the present invention will be described below in detail with reference to the drawings. (First Embodiment) FIG. 1 is a functional block diagram showing an embodiment of an information presentation apparatus to which the present invention is applied. The information presentation device 1 is formed by a database 17, a control pattern file 18, a speech recognition dictionary 19, and a predetermined program that is read and executed by the computer device built in the computer device or an external storage device. , Presentation request input unit 11, data acquisition unit 1
2. It is configured to include an information presentation processing unit 13, a voice input unit 14, a voice recognition unit 15, and a voice control unit 16.

The above program is usually stored in an internal storage device or an external storage device of the computer device, and is read and executed as needed. However, a recording medium separable from the computer device, for example, a CD-ROM. RO
It may be stored in M, FD, or the like, and may be installed in the above-mentioned internal storage device or external storage device at the time of use and provided for execution at any time. That is, the information presentation device of the present embodiment can be realized by a general-purpose computer device and a recording medium that records a program for forming the functional blocks in a computer-readable form.

The database 17 stores data such as documents and images to be presented (hereinafter referred to as presentation target data). The control pattern file 18 is used to control the information presentation processing unit 13 by voice. , A control pattern registered in advance and control information corresponding to the control pattern. This control pattern file 18
For example, it stores a group of control pattern files created for each predetermined unit in presentation target data such as a page. An example of this control pattern file is shown in FIG.
Shown in

A vocabulary group to be subjected to speech recognition is registered in the speech recognition dictionary 19 in advance. This voice recognition dictionary 1
Numeral 9 stores a speech recognition dictionary group created for each predetermined unit in presentation target data such as a page. FIG. 5 shows an example of the speech recognition dictionary. Note that the control pattern file 18 and the speech recognition dictionary 19 are constructed in association with the presentation target data in the database 17, respectively.

The presentation request input unit 11 receives a presentation request from a user or the like input via an input device (not shown) such as a keyboard or a mouse, and performs input to the data acquisition unit 12.

The data acquisition unit 12 acquires data corresponding to the input presentation request from the database 17 and inputs the data to the information presentation processing unit 13.

The information presentation processing unit 13 includes the data acquisition unit 12
Is presented. The information presentation processing unit 13 is configured using, for example, an interface such as a browser in a Web environment. Further, when the information output from the information presentation processing unit 13 (hereinafter referred to as presentation information) is to be controlled by voice, it is configured to cooperate with the voice input unit 14 to receive voice input.

The voice input unit 14 receives a voice from a user or the like input through an input device such as a microphone (not shown) and inputs the voice to the voice recognition unit 15. The voice recognition unit 15 performs a recognition process on the voice input from the voice input unit 14 based on the voice recognition dictionary 19. A recognition result when the voice is recognized (hereinafter, a recognized vocabulary) is input to the voice control unit 16. The voice control unit 16 controls presentation information in the information presentation processing unit 13 based on the recognized vocabulary and the control pattern file 18, that is, performs voice recognition control (hereinafter, voice control).

Next, an information presenting method using the information presenting apparatus 1 will be described according to the procedure of FIG. First, in the presentation request input unit 11, when a presentation request from a user or the like is input (step S101), the presentation request is input to the data acquisition unit 1.
Enter 2 The data acquisition unit 12 acquires data corresponding to the presentation request from the database 17, and inputs the acquired data to the information presentation processing unit 13 to present the data to the user (step S102).

The information presentation processing unit 13 detects whether or not the input data, ie, the presentation information, is a target of voice control. If the input data is a target of voice control (step S103: Yes), the voice input unit 14 is started. On the other hand, if it is not the target of voice control (step S10
3: No), the process returns to step S101, and an input of a new presentation request is accepted.

The voice input unit 14 obtains voice input from a user or the like via an input device such as a microphone (step S104), and inputs the voice to the voice recognition unit 15. The voice recognition unit 15 performs the voice recognition process (Step S105). In this case, the recognition process is performed by referring to a vocabulary group to be subjected to speech recognition corresponding to the presentation information in step S102 from the speech recognition dictionary 19 and collating with the speech, that is, pattern matching. When the voice is recognized (Step S106: Ye
s) The corresponding recognized vocabulary is input to the voice control unit 16. On the other hand, if they do not match (step S106: N
o), a notice to that effect is given by a message or the like as appropriate, and the process returns to step S104 to re-input the voice.

The voice control unit 16 compares the recognized vocabulary after voice recognition with the control pattern file 18 (step S107). If the recognized vocabulary does not match the control pattern file 18 (step S108: No), This is notified appropriately by a message or the like, and step S
Returning to step 104, the voice is re-input. In this case, for example, by appropriately setting a control pattern to be a default value in the control pattern file 18, the recognized vocabulary recognized in steps S105 to S106 does not explicitly exist in the control pattern file 18. Such a case can be dealt with.

On the other hand, while the recognized vocabulary matches the control pattern file 18 (step S108: Yes),
If the recognized vocabulary is other than the control pattern “end” (step S109: No), the voice control unit 16 performs a presentation process on the presentation information of the information presentation processing unit 13 based on the control pattern (step S110). In the presentation process in this case, for example, data corresponding to the control pattern is sequentially acquired from the database 17 by the data acquisition unit 12 and input to the information presentation processing unit 13 to be presented. Alternatively, at the time of acquiring data corresponding to the presentation request in step S102, the data acquisition unit 12 acquires all data related to the data in advance, inputs the data to the information presentation processing unit 13 and holds the data. Control unit 16
Is appropriately configured to perform a presentation process based on the control pattern for the data of the information presentation processing unit 13. If the recognized vocabulary is the control pattern “end” (step S10
9: Yes), the process ends.

Note that the above-mentioned step S109 shows the end processing in the information presenting apparatus 1 for the sake of convenience focusing on the control pattern "end". In practice, however, this control pattern "end" is also a control pattern. It is defined in the file 18 and has the same processing as in step S110.

FIGS. 3 and 4 show the information presentation processing unit 1 of the present invention.
In addition to applying a Web browser to HTML3, Java-based HTML (HyperText Mark
5 is a chart showing a form in a case of being incorporated into an up language) document.

In the HTML document “search.html” in FIG. 3 and the HTML document “faq_top.html” in FIG.
And the description in line 11 represents a declaration for performing voice control. The document is analyzed by the information presentation processing unit 13, that is, the browser, and a program called “RecogCont.class” in 10 lines is executed. This program "RecogCont.cl
In the case of the HTML document “search.html” in FIG. 3, the “ass” is a speech recognition dictionary “Recog1” and a control pattern “Pattern”.
In the case of the HTML document “faq_top.html” in FIG. 4, the voice recognition dictionary “Recog2” and the control pattern “Patt
ern2 "is used as an argument to perform voice control.

FIGS. 5 and 7 show the above program "RecogCon.
"T.class", the speech recognition dictionaries "Recog1" and "R
It is a chart showing an example of "ecog2". In these tables, “recognized vocabulary” indicates a voice to be input, and “output” indicates an output pattern. The “recognized vocabulary” in the table is used to collate with the input voice data, and is described using hiragana for convenience. For example, using a sample of digitized voice data, etc. Is also good. “Output” is an output pattern used for internal processing between the voice recognition unit 15 and the voice control unit 16. For example, a unique character string corresponding to each data of “recognized vocabulary” may be set as appropriate. Good.

FIGS. 6 and 8 show the above program "RecogCon
The control pattern file "Pattern" which is an argument of "t.class"
9 is a chart showing an example of “1” and “Pattern2”. The “control pattern” in these tables represents the recognition vocabulary to be collated, and the action in which the “action” is defined, that is, the control information. "Control pattern" in the figure
Is composed of "output" or a combination of "output" of the recognized vocabulary in the speech recognition dictionary, for example,
“{History | history}” in FIG. 6 indicates a case where the output pattern related to recognition of the speech recognition dictionary is “history” or “history”. “Action” means processing related to presentation corresponding to “control pattern”.
For example, “JUMP“ faq_top.html ”” in FIG.
This indicates that the document is moved to the L document “faq_top.html” and displayed.

FIGS. 9 and 10 are schematic diagrams showing the operation of the information presenting apparatus 1 when using the charts of FIGS.
First, in FIG. 9, the HTML document “search.html” is analyzed, and information of the content indicated by reference numeral 9A is presented on the browser. Since the document is subject to voice control, voice is input at the voice input unit 14. Here, assuming that the voice “go to the first” indicated by reference numeral 9B is input,
The speech recognition unit 15 detects “first” and “go” of the code 9C as recognition vocabulary based on the speech recognition dictionary “Recog1” in FIG. The voice control unit 16 detects a 9D control pattern “Ichiban. * {Go | Fly}” based on the detected recognition vocabulary and the control pattern file “Pattern1” in FIG. 6, and controls this control pattern. Information "JUMP" faq
_top.html ””, the corresponding presentation information is presented to the browser. Reference numeral 9E indicates the action “JUMP“ faq_top.h
tml "", indicating that the presentation information of the browser has changed from 9A to 9E by the input voice.

Similarly, in the case of FIG. 10, the HTML document "fa
“q_top.html” is analyzed, and the information of the content indicated by reference numeral 10A is presented on the browser. The document,
Since the voice is to be controlled, voice is input in the voice input unit 14. Here, assuming that the voice “question about software” shown by reference numeral 10B is input, the voice recognition unit 15 detects “software” of reference numeral 10C as a recognition vocabulary based on the voice recognition dictionary “Recog2” of FIG. .

The voice control unit 16 detects a control pattern "software" of reference numeral 10D based on the detected recognition vocabulary and the control pattern file "Pattern2" in FIG.
"JUMP" software_fa "which is the control information of this control pattern
q.html ””, the corresponding presentation information is presented to the browser. Reference numeral 10E indicates the action “JUMP“ software_fa
q.html "", and the presentation information of the browser is represented by reference numerals 10A to 10E according to the input voice.
It has changed to.

In this embodiment, the database 1
7. Although the control pattern file 18 and the speech recognition dictionary 19 are constructed in a separated form, the present invention is not limited to this embodiment. For example, the control pattern file 18 and the speech recognition dictionary 19 may be appropriately stored in the database 17. May be included.

As described above, the information presentation device 1 of the present embodiment
Then, even when the number of vocabularies to be recognized increases, the recognition vocabulary group is registered and set for each predetermined information unit to construct a speech recognition dictionary, so that the recognition rate and the degree of freedom can be maintained without lowering the practicality. High voice recognition becomes possible. From this, since the recognition vocabulary of the speech recognition dictionary is limited for each predetermined information unit, the recognition accuracy can be maintained at a certain value or more as compared with the conventional method.

Similarly, by setting a control pattern for each predetermined information unit and constructing a control pattern file, the presentation form can be freely controlled for each specific information unit such as a presentation page. it can.

Further, by presenting information by voice recognition, for example, effective assistance can be provided in a case where use of a keyboard is often hindered, such as in a portable terminal, and an environment for presenting a wide range of information can be provided. .

(Second Embodiment) The present invention relates to a system for browsing information by voice recognition for a large amount of digitized information distributed through a public network such as the Internet as a communication line, for example, the information presenting apparatus. It is also possible to implement the present invention in the form of an information browsing system in which a client functioning as an information providing device and an information providing server functioning as an information providing device are deployed.

The client in this case is positioned as, for example, an information acquisition device for a plurality of large-scale databases on the Internet environment. As an example of the configuration, a communication control unit that performs communication with an information providing server via a public network is provided.
It has the same functional blocks as those described above, a presentation request input unit 11, a data acquisition unit 12, an information presentation processing unit 13, a speech input unit 14, a speech recognition unit 15, and a speech control unit 16.

On the other hand, the information providing server stores the database 1 in the computer device or in an external storage device.
7. The same database as the control pattern file 18 and the voice recognition dictionary 19, the control pattern file, and the voice recognition dictionary are constructed, and the communication control unit that communicates with the client is provided.

The difference between the client and the information presenting apparatus 1 is that the client includes a known communication control unit for performing communication control. Configuration information is input to the data acquisition unit 12 via the communication control unit. Similarly, a control pattern file and a voice recognition dictionary corresponding to the obtained digitized information are obtained from the information providing server via the communication control unit, and the obtained control pattern file and the voice recognition dictionary are stored in the client. By doing so, substitution is possible, and it is possible to obtain the same effect as that of the information presentation device 1.

In this information browsing system, it is possible to easily change the speech recognition dictionary and the control pattern file from the information provider side, that is, the information providing server, which is impossible because the conventional system depends on the client environment. And this is reflected in the voice control processing in the client.

[0046]

As is apparent from the above description, according to the present invention, there is an effect that the presentation form of information to be presented in a browser or the like can be flexibly controlled by voice.
In addition, by applying the present invention to a system in which the information providing side and the information obtaining side are separated from each other, the information providing side can easily configure the information in consideration of the presentation form, and the information obtaining side can freely use speech. An environment with high degree and operability can be provided.

[Brief description of the drawings]

FIG. 1 is a functional block diagram of an information presentation device according to an embodiment of the present invention.

FIG. 2 is a processing procedure diagram in the information presentation device.

FIG. 3 is an explanatory diagram showing an example of an HTML document incorporating a voice control process.

FIG. 4 is an explanatory diagram showing an example of an HTML document incorporating a voice control process.

FIG. 5 is an explanatory diagram showing an example of a speech recognition dictionary.

FIG. 6 is an explanatory diagram showing an example of a speech recognition dictionary.

FIG. 7 is an explanatory diagram showing an example of a control pattern file.

FIG. 8 is an explanatory diagram showing an example of a control pattern file.

FIG. 9 is a schematic diagram illustrating voice recognition control in the information presentation device.

FIG. 10 is a schematic diagram illustrating voice recognition control in the information presentation device.

FIG. 11 is a functional block diagram of a conventional information browsing system.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Information presentation apparatus 11 Presentation request input part 12 Data acquisition part 13 Information presentation processing part 14 Voice input part 15 Voice recognition part 16 Voice control part 17 Database 18 Control pattern file 19 Voice recognition dictionary 30 Information providing server 40 Client 110 Information browsing system 111 speech recognition unit 112 speech recognition dictionary 113 speech recognition control unit 114 browser

Claims (9)

[Claims] 1. An information presentation method using a computer device, wherein a presentation form of digitized information to be presented is changed by voice, wherein a recognition vocabulary group to be subjected to voice recognition and the recognition vocabulary group correspond to the recognition vocabulary group. A presentation control information group relating to the presentation of the digitized information is stored in advance for each presentation form of the digitized information, and a recognition vocabulary to be collated with the input speech is detected from the recognition vocabulary group. An information presentation method, comprising: detecting presentation control information to be matched with the presentation control information group; and presenting the corresponding digitized information in a presentation form based on the detected presentation control information. 2. An apparatus for controlling a presentation form of digitized information to be presented by speech, comprising: a speech recognition dictionary in which recognition vocabulary groups to be subjected to speech recognition are registered in advance for each presentation form; A control pattern file in which presentation control information related to the presentation of the digitized information corresponding to a vocabulary group is set in advance for each of the presentation forms, and the digitized information corresponding to an information presentation request from an operator is presented. Means for detecting whether or not the digitized information is a presentation control target by voice; and when the digitized information to be subjected to the presentation control by voice is detected, a voice from the operator is acquired and collated with the voice. Voice recognition means for detecting a recognition vocabulary to be detected from the voice recognition dictionary and performing voice recognition; and detecting, from the control pattern file, presentation control information to be collated with the detected recognition vocabulary. To together, the information presentation device being characterized in that and a presenting means for presenting the corresponding electronic information presentation form based on the presentation control information. 3. The speech recognition dictionary is constructed so as to find out, as a recognition vocabulary, unique data corresponding to a plurality of speech recognition data having the same data format that can be collated with a voice from the operator. 3. The method according to claim 2, wherein
Information presentation device as described. 4. The speech recognition dictionary is constructed so as to find out, as a recognition vocabulary, a unique character string corresponding to a plurality of speech recognition data having the same data format that can be collated with a voice from the operator. 3. The method according to claim 2, wherein
Information presentation device as described. 5. The control pattern file according to claim 2, wherein the control pattern file is constructed to search for the presentation control information corresponding to the control pattern including the recognized vocabulary or a combination of the recognized vocabulary. Information presentation device. 6. The information presentation apparatus according to claim 2, wherein the speech recognition dictionary and the control pattern file are constructed by being included in a database storing the digitized information. 7. The system according to claim 2, wherein the presenting unit is configured to present the digitized information based on an information unit corresponding to a presentation region of the digitized information.
Information presentation device as described. 8. The computerized information, wherein the computerized information to be presented, the speech recognition vocabulary group to be subjected to speech recognition, and the presentation control information group related to the presentation of the computerized information corresponding to the speech recognition vocabulary group are represented by A first device, which is created and held in advance for each presentation mode, and a second device, which presents the digitized information corresponding to an information presentation request from an operator, is connected in a two-way communication manner. Means for acquiring and presenting the digitized information corresponding to the information presentation request from the operator from the first device, and detecting whether or not the digitized information is a presentation control target by voice; and When the digitized information to be subjected to the presentation control by the voice is detected, a voice from the operator is obtained, and the recognition vocabulary group corresponding to the digitized information is obtained from the first device, and Match with audio Means for detecting a linguistic vocabulary; detecting presentation control information for acquiring the presentation control information group corresponding to the digitized information to be subjected to the presentation control from the first device and collating with the detected recognition vocabulary; ,
Presentation means for acquiring and presenting the corresponding digitized information from the first device in a presentation form based on the presentation control information;
An information presentation system, comprising: controlling a presentation form of the digitized information to be presented by voice. 9. The information presentation apparatus according to claim 8, wherein said presentation means is configured to present said digitized information based on an information unit corresponding to a presentation area of a predetermined Web browser. system.